tzl
tzl
PostsComments
MModular
Created by tzl on 5/14/2024 in #questions
Why is this in-place operation not modifying the Python object in-place?
from python import Python

def do_numpy_stuff(ar: PythonObject) -> PythonObject:
    ar.__iadd__(3)
    print("inside function:\n", ar)

    return ar


fn main() raises:
    var np = Python.import_module("numpy")
    var ar = np.arange(15).reshape(3, 5)
    print(ar)

    print("do_numpy_stuff:")
    do_numpy_stuff(ar)
    print("outside function:\n", ar)
from python import Python

def do_numpy_stuff(ar: PythonObject) -> PythonObject:
    ar.__iadd__(3)
    print("inside function:\n", ar)

    return ar


fn main() raises:
    var np = Python.import_module("numpy")
    var ar = np.arange(15).reshape(3, 5)
    print(ar)

    print("do_numpy_stuff:")
    do_numpy_stuff(ar)
    print("outside function:\n", ar)
I would expect the outside function ar print to print the modified ar, because the ar is passed as object reference. But I got this: 
> mojo do_numpy_stuff.mojo 
[[ 0  1  2  3  4]
 [ 5  6  7  8  9]
 [10 11 12 13 14]]
do_numpy_stuff:
inside function:
 [[ 3  4  5  6  7]
 [ 8  9 10 11 12]
 [13 14 15 16 17]]
outside function:
 [[ 0  1  2  3  4]
 [ 5  6  7  8  9]
 [10 11 12 13 14]]
> mojo do_numpy_stuff.mojo 
[[ 0  1  2  3  4]
 [ 5  6  7  8  9]
 [10 11 12 13 14]]
do_numpy_stuff:
inside function:
 [[ 3  4  5  6  7]
 [ 8  9 10 11 12]
 [13 14 15 16 17]]
outside function:
 [[ 0  1  2  3  4]
 [ 5  6  7  8  9]
 [10 11 12 13 14]]
8 replies